Recently, a mobile communication terminal provides communication functions such as telephone or text message transmission services as well as various additional functions utilizing personal information such as mobile banking, etc. Accordingly, the necessity of locking devices of the mobile communication terminal is considered more important.
Most existing locking devices applied to the mobile communication terminal used a conventional method of using a password. For example, the conventional method was to apply a locking device to the telephone function, additional function, or international telephone function, etc., and to input the corresponding password in order to use the corresponding function.
However, this method caused inconveniences that it became useless when the password is revealed, the password should be periodically changed to secure stability, a user should memorize the password, etc.
Thus, recently, in order to supplement this method and improve locking effect, terminals with locking devices using fingerprint recognition have been developed in earnest.
In order to mount this fingerprint recognition device on mobile communication terminals which are manufactured compactly, the size of its sensor should be minimized. Accordingly, a fingerprint recognition sensor with ‘a sliding method’ which recognizes a fingerprint by swiping a finger on a bar-shaped sensor which is long from side to side has been developed.
FIGS. 1 and 2 are views illustrating an operation of a fingerprint recognition sensor with a conventional sliding method.
Referring to FIGS. 1 and 2, when a user touches a fingerprint sensor 10 with his finger and moves his finger with the sliding method, partial short images P1˜P4 are acquired continuously.
A fingerprint includes various minutiae. Specifically, the fingerprint includes an ending where a ridge is cut off, bifurcation where two ridges are met, etc. The minutiae vary depending on people, so the minutiae derived from short images P1˜P4 also vary depending on people.
As illustrated in FIG. 2, the conventional method of fingerprint recognition utilizes a method of selecting minutiae from short images P1˜P4, matching minutiae derived from each short image P1˜P4 to make one minutiae template, establishing database for the position and number of minutiae, and performing comparison and analysis.
The fingerprint database is made by fingerprint registration by a specific user, and a method for registering a fingerprint is performed in the same manner as the method for fingerprint recognition.
This conventional method for fingerprint registration and fingerprint recognition utilizes a method of matching minutiae of the fingerprint, so matching information needs to be simple, thereby resulting in errors in fingerprint recognition. Additionally, when the registered fingerprint does not match with the recognized fingerprint in terms of direction, etc., there is a problem that the fingerprint will not be matched. Also, the fingerprint is sensed in a small-sized bar-shaped sensor 10, so an entire image of the fingerprint cannot be acquired, and accordingly, there is no doubt that recognition accuracy decreases when only part of the fingerprint image is inputted.